VO-BB - 20 YEARS OLD!

[Quicksilver]

Question for you audio experts..

Is there anyway to tell if a WAV is just an MP3 re-saved as a WAV? Outside of being able to hear it?

[Bish]

It's all about the ears. I honestly don't think I could tell if you'd taken a 320kbps mono mp3 and converted it to a 44.1 16-bit wave file. There are those who will tell you they can hear the difference between the different mp3 sampling rates or a 16 or 24 bit wav file. Personally, I think this is quite possible for someone with much better ears (e.g. much younger ears) than mine. I had a long conversation with an engineer who insisted he could hear the artifacts in a 256k mono mp3. Personally, I am skeptical, but I am not a good judge. Noted that you are talking of 320kbps... 128kbps mp3 files can be grungy... of that there is no doubt.

The only way to test this is to do a blind test... the original wav, and one converted to 320kbps mp3 then re-converted back to wav.

Edited to add: This is purely about listening... uncompressed wav/aiff files are an absolute must when doing further production work where additional processing is going to happen... it's important to have as clean and accurate a file as possible.
_________________
Bish a.k.a. Bish
Smoke me a kipper... I'll be back for breakfast.
I will not feed the trolls... I will not feed the trolls... I will not feed the trolls... I will not feed the trolls.

[Kim Fuller]

Thanks for asking this - I was wondering about this recently too. Thought there might be some degradation of the file...

[Quicksilver]

Thanks Bish.

Just curious, why is lossless so important when doing further processing if most (or nearly all, as we suspect) can't hear the difference from a 320?

[Jason Huggins]

There is definitely degradation. You could probably tell in headphones if you were to A/B files...maybe not at higher bitrates. I can DEFINITELY tell by looking at the Spectral view in Audition. It is easy to tell an MP3 from a WAV file because there is missing information. Especially above 16k.

[Bish]

Jason is right... I slightly mis-read the original post (not enough coffee)... the whole "outside of being able to hear it?" slipped past me. Either way it's an interesting discussion.

The moral of this story is to drink more coffee before placing fingers on the keyboard
_________________
Bish a.k.a. Bish
Smoke me a kipper... I'll be back for breakfast.
I will not feed the trolls... I will not feed the trolls... I will not feed the trolls... I will not feed the trolls.

[vkuehn]

This is a topic that I learned the hard way in photography before it finally grabbed my full attention in audio. Many of us began our digial photography with a camera that only saved in .jpg. YOu open the file in photoshop which converts the file to the photographers equivalent of a wav file, 24 or 32 bit. You change a few colors here, So some things over there, and then you get called to the family dinner table so you save back to .jpg. I did a wedding photograph for my daughter ten years ago. I have no idea how many times I opened the file (a conversaion) and restored the file back to the hard drive (another conversion).

Then I bought a camera that would save photographs as .raw! And then after each little editing session, store the file away as a .tiff or Photoshops internal equivalent. OH MY WORD!! I picked up hints here and there in discussion groups and photo magazines and finally one day it hit me: We have a parallel in audio!!!

So now I record in at least 24-bit and if the hardware and software permit, 32-bit. I always convert and KEEP the working file in 32-bit. Only when I run out of tweaks to make do I even think about "dumbing the file down" to what the customer wants. (And always keep a copy of the 32-bit file in case you need to correct or modify the project later.)

We have know way of know in advance which clients can hear the different and which one cannot hear the difference, so the smart money probably says: While editing, assume they ALL can hear the difference.

[vkuehn]

I don't want to create a thread-jack... just continue this topic but pose another question that fits in.

Let's take the ACX instructions for producers. Record at 44.1, 16 bit. Convert to mp3... what is it 192 kbps?

So like a good Boy Scout I took my master files (wav, 32 bit) and converted down to 16 bit.... then converted that file to mp3. They will never know that I recorded at a better bit rate, or maybe even recorded at 48k or 96k for recording and editing purity. But then ALWAYS convert to 44.1/16 BEFORE conversion to mp3.

And then in a fit of inspiration one day, I said: "Hoss, you can take the 48 or 88.2 or 96k file at 32 bits and simply convert to MP3, 192 kbps. No one will ever know if you did the intermediate step of down-converting to 44.1/16 before the conversion to mp3.

So, am I on the money, or do the quality control folks at production houses have the ability to do an autopsy on an MP3 file and see all the genealogy?

Can the mp3 converter do a more pristine conversion if the original file has more quality before throwing away stuff in the conversion process?

[Ed Fisher]

To paraphrase.....someone from this site:

"if you Ears can't tell the difference...THEN...there is no difference."

And...unless someone had "Teenage Ears" and heard the Wav and the MP3 right next to each other back and forth a few times...it's still doubtful that they could tell the difference. (unless you used some ridiculously low sample rate on the mp3..in which case the answer is there is ABSOLUTELY a difference.)
_________________
"I reserve the right to be completely wrong."

[Eddie Eagle]

It's off to Coffee Detention for you Bish!
Double Espressos for a week young man.

[vkuehn]